Autistic disorder is a severe disorder of cognitive, language, and social development. Several studies have indicated genetic predisposition to the development of autistic disorder in children. Subjects with suspected autistic disorder will be diagnosed by use of the parent interview, Autism Diagnostic Interview-Revised (ADI-R) and clinical DSM-IV diagnosis based on interview of the child with the Autism Diagnostic Observation Schedule (ADOS) and review of the transcript of the ADI-R parent interview without access to the ADI-R scoring. Subjects will he excluded from genetic analysis if their mental age is less than 18 months or if they have Fragile X syndrome, phenylketonuria, tuberous sclerosis, or other identifiable chromosomal disorders. These subjects with autistic disorder will be studied to determine if there is an association between the primary candidate genes, serotonin 5-HT2A receptor, 5-HT7 receptor, 5-HT transporter, and dopamine D2 receptor and autistic disorder. In addition, the following secondary candidate genes will he studied: (5-HT1A, 5-HT1B, 5-HT1D, 5-HT1E, 5-HT1F, 5-HT2B, 5-HT2C, 5-HT3, 5-HT4, 5-HT5A, 5-HT6), other dopamine receptors (D1, D3, D4, D5), enzymes involved in serotonin and/or dopamine metabolism or transport (tryptophan hydroxylase, tyrosine hydroxylase, dopamine-beta-hydroxylase, dopamine transporter, aromatiC amino acid decarboxylase, monoamine oxidase (MAO)-A and MAO-B). C-Harvey-Ras-1 (HRAS) will be studied because of a previous association found in autistic disorder without using intrafamilial controls. The haplotype-based Haplotype Relative Risk (HRR) method will be used to assess the overall evidence for association between alleles at the above candidate genes and autistic disorder. If a significant association between alleles at any of these candidate genes and autistic disorder is found, the Transmission/Disequilibrium Test (TDT) will be used to determine whether the observed association is due to linkage between the marker and susceptibility to autistic disorder. The statistical methods have been chosen to reduce the change of a false positive result due to population stratification.